How to delete or ignore the rows with duplicate values in time

I have a 26400x2 table with time and numbers representing satellite IDs for every second (as in the picture, the file is too large to upload). I need to get the number of satellites in time (which I want to plot later). This should simply be the height of each vector containing 1 second. However, there are some duplicate sat IDs in each second, and I don't know how to remove them. I tried the following:

t1 = raw.UTCTime(1);
t2 = raw.UTCTime(end);
t = (t1:seconds(1):t2)';

num_sats_gps = zeros(length(t),1);
for i = 1:length(t)
    idx = gps.UTCTime == t(i);
    uniq = unique(gps.Svid(idx));
    num_sats_gps(i,1) = height(uniq);
end

but it doesn't work like it's supposed to, the result gives me vector with 0 values. Is there any other approach to this? Any answer would be highly appreciated!

Answer 1

[Yash]

There might be an issue with how you're indexing the table or filtering duplicates, other than that, your code seems right to me.

As an alternative, you can use the 'splitapply' function. Take the help of the below mentioned code.

% Assuming your table is named 'raw' with columns 'UTCTime' and 'Svid'

t1 = raw.UTCTime(1);

t2 = raw.UTCTime(end);

t = (t1:seconds(1):t2)';

%Storing the resulting count in num_sats_gps

num_sats_gps = splitapply(@(x) numel(unique(x)), raw.Svid, discretize(raw.UTCTime, t));

%Using 'discretize' function to divide the time range into one-second intervals

The splitapply function applies the anonymous function @(x) numel(unique(x)) to each group of satellite IDs (x) associated with a specific time interval. It counts the number of unique satellite IDs in each group.

Hope this helps.

[Simona Blaskova]

Thank you so much, your solution works perfectly! I think the problem was with time, which needed to be discretized.